Research ArticleBiofilm Microenvironment-Responsive Nanotheranostics forDual-Mode Imaging and Hypoxia-Relief-Enhanced PhotodynamicTherapy of Bacterial Infections

Weijun Xiu,1 Siyu Gan,1 Qirui Wen,1 Qiu Qiu,1 Sulai Dai,1 Heng Dong ,2 Qiang Li,2

Lihui Yuwen ,1 Lixing Weng,3 Zhaogang Teng,4 Yongbin Mou ,2 and Lianhui Wang 1

1Key Laboratory for Organic Electronics and Information Displays & Jiangsu Key Laboratory for Biosensors, Institute of AdvancedMaterials (IAM), Jiangsu National Synergetic Innovation Centre for Advanced Materials (SICAM), Nanjing University of Postsand Telecommunications, Nanjing 210023, China2Department of Oral Implantology, Nanjing Stomatological Hospital, School of Medicine, Nanjing University, Nanjing 210023, China3School of Geography and Biological Information, Nanjing University of Posts and Telecommunications, Nanjing 210023, China4Department of Medical Imaging, Jinling Hospital, School of Medicine, Nanjing University, Nanjing 210002, China

Correspondence should be addressed to Lihui Yuwen; iamlhyuwen@njupt.edu.cn and Lianhui Wang; iamlhwang@njupt.edu.cn

Received 10 January 2020; Accepted 5 March 2020; Published 24 March 2020

Copyright © 2020 Weijun Xiu et al. Exclusive Licensee Science and Technology Review Publishing House. Distributed under aCreative Commons Attribution License (CC BY 4.0).

The formation of bacterial biofilms closely associates with infectious diseases. Until now, precise diagnosis and effective treatmentof bacterial biofilm infections are still in great need. Herein, a novel multifunctional theranostic nanoplatform based on MnO2nanosheets (MnO2 NSs) has been designed to achieve pH-responsive dual-mode imaging and hypoxia-relief-enhancedantimicrobial photodynamic therapy (aPDT) of bacterial biofilm infections. In this study, MnO2 NSs were modified withbovine serum albumin (BSA) and polyethylene glycol (PEG) and then loaded with chlorin e6 (Ce6) as photosensitizer to formMnO2-BSA/PEG-Ce6 nanosheets (MBP-Ce6 NSs). After being delivered into the bacterial biofilm-infected tissues, the MBP-Ce6NSs could be decomposed in acidic biofilm microenvironment and release Ce6 with Mn2+, which subsequently activate bothfluorescence (FL) and magnetic resonance (MR) signals for effective dual-mode FL/MR imaging of bacterial biofilm infections.Meanwhile, MnO2 could catalyze the decomposing of H2O2 in biofilm-infected tissues into O2 and relieve the hypoxic conditionof biofilm, which significantly enhances the efficacy of aPDT. An in vitro study showed that MBP-Ce6 NSs could significantlyreduce the number of methicillin-resistant Staphylococcus aureus (MRSA) in biofilms after 635 nm laser irradiation. Guided byFL/MR imaging, MRSA biofilm-infected mice can be efficiently treated by MBP-Ce6 NSs-based aPDT. Overall, MBP-Ce6 NSsnot only possess biofilm microenvironment-responsive dual-mode FL/MR imaging ability but also have significantly enhancedaPDT efficacy by relieving the hypoxia habitat of biofilm, which provides a promising theranostic nanoplatform for bacterialbiofilm infections.

1. Introduction

Bacterial infection is a prominent threat for human health.Numerous bacterial infections, including dental caries, cysticfibrosis, pneumonia, otitis media, and especially chronicwounds, usually relate with the formation of bacterial bio-films [1–5]. Biofilm is the aggregated bacterial populationswhich are usually encapsulated in the extracellular polymericsubstances (EPS) and attach to the living or inert surfaces[6, 7]. The compact EPS of bacterial biofilm not only

protect themselves from the attack of host immune systembut also cause serious antibiotic resistance, which bringgreat challenge to eradicate bacterial biofilm infections[8, 9]. Specific and sensitive diagnosis of bacterial biofilminfections is essential to effectively treat these diseases[10–12]. Traditional diagnostic methods, such as culturemethod, biochemistry identification, and polymerase chainreaction (PCR), have been widely used in bacterial biofilminfections diagnosis.However, these ex vivomethods generallyneed invasive tissue ablation, time-consuming procedures,

AAASResearchVolume 2020, Article ID 9426453, 15 pageshttps://doi.org/10.34133/2020/9426453

and have low sensitivity [13–15]. Noninvasive and effectivedetection methods of bacterial biofilm infections in vivo arehighly desired.

Recently, various imaging technologies have been devel-oped for the detection of bacterial infections. Considered asa sensitive method for disease diagnosis, fluorescence imag-ing (FLI) has received great attention [15, 16]. Numerousbacteria-targeting fluorescent imaging probes have beendesigned for FLI of bacterial infections, such as maltodextrin,vancomycin, and N-acetylmuramic acids functionalized dyes[17–19]. Nevertheless, the light scattering and absorptionin biological tissues usually cause poor penetration of FLI[20, 21]. In comparison, magnetic resonance imaging (MRI)has the advantages of deep tissue penetration and high spatialresolution [22]. Previous work has reported the use of MRItechnique to monitor the bacterial infections [23]. However,relatively low sensitivity of MRI limits its application forbacterial infections [22]. Currently, most imaging probes withbacterial detection ability only work in single mode, whichusually limits their use due to the intrinsic shortcomings ofeach imaging mode. Moreover, these imaging probes men-tioned above were designed to target planktonic bacteriarather than bacterial biofilms. The dense EPS matrix of bio-film may prevent the penetration of imaging agents and limittheir application for bacterial biofilm infections. In addition,currently used imaging probes for bacterial infections usuallywork in “always-on” mode that may cause significant back-ground signal and reduce their sensitivity. Therefore, FL/MRdual-mode imaging probes with bacterial biofilm responsiveand activable ability are very promising.

Due to the encapsulation of EPS and fermentation ofbacteria inside biofilm, the biofilm microenvironment has

several characteristics, such as the lack of oxygen, low pH,and high level of H2O2 compared with healthy tissues[24–29]. Hence, biofilm microenvironment is an ideal targetfor specific diagnosis and treatment of biofilm-related infec-tion diseases [29–31]. Previous studies have demonstratedthe use of charge-switchable nanozymes to respond theacidic biofilm microenvironment through bioorthogonalreaction to activate profluorophores and realize specificfluorescence detection of biofilm in vitro [24]. Moreover,pH-responsive Au nanoparticles and micellar nanocarriershave also been designed to combat bacterial biofilm infec-tions [32, 33]. Although these studies are promising, theyusually work in single mode (diagnosis or treatment) andlack the ability of simultaneous diagnosis and therapy. Tothe best of our knowledge, biofilm microenvironment-responsive nanotheranostics that combine both activabledual-mode FL/MR imaging and treatment abilities to combatbiofilm infections are rarely reported.

Here, we designed a biofilm microenvironment-responsive theranostic nanoagent (MBP-Ce6 NSs) for dual-mode FL/MR imaging and enhanced aPDT of MRSA biofilminfections. As Scheme 1 shows, MnO2 nanosheets (MnO2NSs) were modified with biocompatible BSA and PEG toform MnO2-BSA/PEG NSs (MBP NSs). Then, a photosensi-tizer Ce6 was further loaded to prepare MBP-Ce6 NSs. Thefluorescence of Ce6 could be quenched after loading on thesurface of MnO2 NSs. In acidic biofilm microenvironment,MnO2 NSs were decomposed and generate Mn2+ ions,which can be used for biofilm responsive T1-weighted MRI[21, 33, 34]. Meanwhile, the decomposition of MnO2 willalso result in the release of Ce6 from the MBP-Ce6 NSs,restoration of the fluorescence of Ce6, and realization of

H2O2

FLI

MRIMn2+

Ce6

O2

H2O2

BSA

PEG

Ce6

MnO2 NSs MBP NSs MBP-Ce6 NSs

H+

Laser

H+

MRSA

Dead MRSA

Scheme 1: The preparation of MnO2-BAS/PEG-Ce6 nanosheets (MBP-Ce6 NSs) and their use for responsive FL/MR imaging and enhancedaPDT of bacterial biofilm infections.

2 Research

the biofilm-activable fluorescence imaging. Moreover, thereleased Ce6 can also serve as the photosensitizer foraPDT and generate singlet oxygen (1O2) under light irradia-tion. 1O2 can effectively damage the vital biomacromolecules(DNA, protein, etc.) and has been used for the treatment ofcancer and bacterial infection [28, 34, 35]. Since it is difficultfor bacteria to develop resistance to 1O2, aPDT is regardedas an effective approach for the drug-resistant bacterial infec-tion [32, 36]. However, the therapeutic efficacy of aPDTwould be restricted by hypoxia condition in biofilm-infectedtissues [4, 26]. Due to the high level H2O2 in biofilm-infected tissues, MnO2 NSs can convert the H2O2 into O2and enhance O2-dependent aPDT. MBP-Ce6 NSs weresuccessfully used for acid-activated dual-mode imaging andhypoxia relieving aPDT of MRSA biofilm-infected mice.

2. Results and Discussion

2.1. Material Preparation and Characterization. MnO2 NSswere synthesized according to the reported method [37].Manganese (II) chloride reacted with H2O2 and tetramethy-lammonium hydroxide (TMA·OH) to form aggregatedMnO2 NSs (Figure S1, Supporting Information). MnO2NSs were prepared by ultrasonication-assisted exfoliationof the as-prepared MnO2 NSs and followed by gradientcentrifugation. Transmission electron microscopy (TEM)images (Figure 1(a); Figures S2(a) and (d), SupportingInformation) show that the as-prepared MnO2 NSs haveuniform sheet-like morphology and small average size(<100nm). As the selected area electron diffraction (SAED)pattern (Figure S3(a), Supporting Information) shows, MnO2NSs have a polycrystalline structure. A high-resolution TEM(HRTEM) image (Figure 1(b)) shows the distinct crystallattice spacing about 0.21nm, which can be ascribed to the(202) plane of MnO2 [20, 37]. A high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) image and energy dispersive spectroscopy (EDS)elemental mapping images (Figures S3(b), (c), (d), and (e),Supporting Information) show homogeneous elementdistribution in MnO2 NSs. Atomic force microscopy (AFM)images (Figure 1(c); Figures S4(a) and (d), SupportingInformation) reveal that the average thickness of MnO2NSs is about 1.08 nm, which proves the single-layerstructure of MnO2 NSs [37]. To enhance the stability andbiocompatibility of MnO2 NSs, BSA and PEG3000-NH2were used to modify the surface of MnO2 NSs through thevan der Waals interaction and Mn-N coordinate bondingto form MBP NSs [20, 38]. Next, Ce6 was further loaded onthe surface of MBP NSs by similar noncovalent interactions[39]. As the TEM images (Figure 1(e) and (i); Figures S2(b),(c), (e), and (f), Supporting Information) and HRTEMimages (Figure 1(f) and (j)) show, the morphology, size,and crystal structure of the MBP NSs and MBP-Ce6 NSshave no obvious change compared with MnO2 NSs. Thethickness of MBP NSs increased to about 5.07 nm(Figure 1(g) and (h); Figures S4(b) and (e), SupportingInformation) due to the existence of BSA and PEG.After loading with Ce6, the thickness of MBP-Ce6 NSs

further increased to about 5.36 nm (Figure 1(k) and (l);Figures S4(c) and (f), Supporting Information).

The crystal structure of MnO2 NSs was determined byX-ray diffraction (XRD). Based on the XRD pattern of MnO2NSs shown in Figure 1(m), three diffraction peaks, located at8.7°, 18.8°, and 37.0°, can be assigned to the (001), (002), and(100) planes, respectively [37]. The Fourier transform infra-red spectroscopy (FT-IR) and X-ray photoelectron spectros-copy (XPS) were also carried out to further investigate thecompositions of MBP-Ce6 NSs. As shown in Figure 1(n),the IR absorption bands of MnO2 NSs around 500 cm-1

belong to the Mn-O stretching vibrations [37]. The IRabsorption bands of MBP NSs at 2926 cm−1 and 2855 cm−1

can be ascribed to the C-H stretching vibrations from thePEG [40]. The IR adsorption bands of MBP-Ce6 NSs near1650 cm−1 and 1590 cm-1 can be assigned to the C=O bend-ing vibrations from BSA and the N-H bending vibrationsfrom Ce6, respectively [40, 41]. The XPS survey spectrum(Figure 1(o)) of MnO2 NSs shows the binding energy peakslocated at 654 eV and 642 eV, which belong to Mn (IV)2p1/2 and Mn (IV) 2p3/2 of MnO2 [42]. For the MBP-Ce6NSs, the intensity of Mn 2p peaks decreases, while theintensity of N 1s (400 eV) and C 1s (285 eV) peaks increases,indicating the presence of BSA, PEG, and Ce6 on the surfaceof MnO2 NSs. As Figure S5(a) (Supporting Information)shows, the hydrodynamic diameter of MBP-Ce6 NSs islarger than that of MnO2 and MBP NSs, suggesting thesuccessful surface modification. The zeta potential ofMnO2 NSs is about -7.62mV (Figure S5b, SupportingInformation) [37]. After surface modification with BSA andPEG, it increased to -2.05mV because of the existence ofelectroneutral PEG. While further loaded with Ce6, the zetapotential of MBP-Ce6 NSs decreased to about -35.70mVowing to the negative charge of Ce6 [41].

Ce6 loading and fluorescence quenching abilities of MBPNSs were also investigated. As shown in Figure 2(a), thefluorescence intensity of Ce6 (20μg/mL) nearly diminishedafter being mixed with MBP NSs (MnO2: 48μg/mL) dueto the quenching of MnO2 NS [43, 44]. The ultraviolet-visible-near infrared (UV-Vis-NIR) absorption spectrum(Figure 2(b)) of MBP-Ce6 NSs shows that the absorbanceof the characteristic peaks for Ce6 near 400 and 650nm ismuch higher than that of MBP NSs and MnO2 NSs, sug-gesting the successful loading of Ce6 on MBP NSs [45]. Asillustrated in Figure 2(c), the loading efficiency of Ce6reaches as high as 67.5% at the feeding ratio of 6 : 1(mCe6 :mMBP), while the corresponding encapsulation effi-ciency is about 22.5%, indicating good Ce6-loading capac-ity of MBP NSs.

Since MnO2 nanostructures could be decomposed inacidic aqueous solutions and release Mn2+, MnO2 NSs havebeen used as pH-responsive drug delivery platforms [46].The pH-responsive release of Ce6 by MBP-Ce6 NSs was thenstudied. The morphology change of MBP-Ce6 NSs afterbeing incubated in phosphate-buffered saline (PBS) with dif-ferent pH values was studied by using TEM. As shown inFigure 2(d), the MBP-Ce6 NSs gradually broke up in acidiccondition (pH = 5:0) after 4 h incubation, while those in neu-tral PBS (pH = 7:4) still kept intact morphology even after

3Research

24 h storage. The decomposition of MBP-Ce6 NSs was alsostudied by using the UV-Vis-NIR absorption spectroscopy.As indicated in Figure 2(e), the degradation of MBP-Ce6NSs was less than 20% in PBS at pH = 7:4 after 24 h storageat room temperature. In comparison, nearly all the MBP-Ce6 NSs were decomposed in acidic PBS at pH = 5:0 and6:0. The acid-induced decomposition of MBP-Ce6 NSs couldtrigger the fast release of Ce6. The amount of released Ce6

and the fluorescence recovery were studied by the UV-Vis-NIR absorption spectroscopy and fluorescence spectroscopy,respectively. As illustrated in Figure 2(f), the release ofCe6 is much faster in acidic conditions than in the neutralcondition. In acidic PBS (pH = 5:0), the released Ce6reached 52.64% after 24h incubation, while that in neutralPBS (pH = 7:4) was only 18.69%. Figure S6 (SupportingInformation) further indicates that the fluorescence of

1.5

0.5

1

00 60

Distance (nm)

Hei

ght (

nm)

120

6.0

(a) (b) (c)

(d)

(e) (f) (g)

(h)

(i) (j) (k)

(m) (n)

(o)

(l)

0

3.0

0 60Distance (nm)

Hei

ght (

nm)

120

6.0

0

3.0

0

10 30

002

100

001

2𝜃 (degree) Wavenumber (cm–1)50 4000 3000 2000 1000 600

Mn 2p2926 cm–1 2855 cm–1

MBP-Ce6

MBP

1650 cm–1 1590 cm–1

500 cm–1O 1s

C 1sN 1s

MnO2

MnO2

MBP-Ce6

Binding energy (eV)400 200

80Distance (nm)

Inte

nsity

(a.u

.)

Inte

nsity

(a.u

.)

Tran

smitt

ance

(%)

Hei

ght (

nm)

160

Figure 1: Characterization of MnO2 NSs, MBP NSs, and MBP-Ce6 NSs. (a) TEM image, (b) HRTEM image, (c) AFM image, and (d) heightprofile of MnO2 NSs. (e) TEM image, (f) HRTEM image, (g) AFM image, and (h) height profile of MBP NSs. (i) TEM image, (j) HRTEMimage, (k) AFM image, and (l) height profile of MBP-Ce6 NSs. (m) XRD pattern of MnO2 NSs. (n) FT-IR spectra of MnO2 NSs, MBPNSs, and MBP-Ce6 NSs. (o) XPS spectra of MnO2 NSs and MBP-Ce6 NSs.

4 Research

Ce6 reached the highest intensity after 12 h storage, andthe release of the Ce6 in acidic conditions is much fasterthan that in neutral condition.

Hypoxia is another typical characteristic of the microen-vironment in bacterial biofilm-infected tissues [47, 48]. Sinceoxygen is an essential factor in the process of PDT, the

200 400 600 800640

(a) (b) (c)

(d)

(e) (f) (g)

660 680Wavelength (nm) Wavelength (nm)

700 720

0 h

pH = 7.4

pH = 5.0

4 h 8 h 16 h 24 h

0 𝜇g/mL8 𝜇g/mL16 𝜇g/mL24 𝜇g/mL32 𝜇g/mL40 𝜇g/mL48 𝜇g/mL

0 𝜇g/mL5 𝜇g/mL10 𝜇g/mL20 𝜇g/mL

740

120 2.0

Ce6

100

80

60

40

20

Load

ing

effici

ency

(%)

Enca

psul

atio

n effi

cien

cy (%

)

0

10075 5

pH = 7.4pH = 6.0pH = 5.0

432

1

0

60

45

30

15

80604020

Reta

ined

MnO

2 (%

)

Rele

ased

Ce6

(%)

Diss

olve

d O

2 (m

g/L)

0

0 5 10 15Time (h) Time (h) Time (s)

20 25 0 5 10 15 20 25 0 150 300 450 600

100

80

60

40

20

00 1 2Feeding ratio (mCe6 : mMBP)

3 4 5 6

MnO2MBPMBP-Ce6

1.0

Abso

rban

ce (a

.u.)

0

60

Fluo

resc

ence

inte

nsity

(a.u

.)

0

pH = 7.4pH = 6.0pH = 5.0

Figure 2: Characterization and drug release of MBP-Ce6 NSs. (a) The fluorescence spectra of Ce6 aqueous solutions (20 μg/mL) after beingmixed with MnO2 NSs (0, 6, 12, 18, 24, 30, 36, and 48μg/mL). (b) The UV-Vis-NIR spectra of Ce6, MnO2 NSs, MBP NSs, andMBP-Ce6 NSs.(c) The loading efficiency and encapsulation efficiency of Ce6 at different feeding ratios of Ce6 and MnO2 NSs. (d) The TEM images of MBP-Ce6 NSs after being stored in PBS at different conditions. The scale bar is 100 nm for all TEM images. (e) The time-dependent degradation ofMBP NSs (40 μg/mL) in PBS at different pH values. (f) Released Ce6 from MBP-Ce6 NSs over time at different conditions. (g) The time-dependent concentration change of dissolved O2 in H2O2 solutions (50 μM) after being mixed with MBP-Ce6 NSs at differentconcentrations (0, 5, 10, and 20μg/mL). Error bars indicate s.d. (n = 3).

5Research

hypoxia condition can significantly limit the therapeutic effi-cacy of PDT. Fortunately, H2O2 usually exists at a relativelyhigh level in bacteria-infected tissues and can be used asa potential oxygen source [28, 49–51]. Therefore, theperoxidase-like ability of MBP-Ce6 NSs was investigated.After MBP-Ce6 NSs were dispersed in H2O2 aqueoussolution (50μM), the concentration of dissolved O2 in themixture changed according to the concentration of MnO2(Figure 2(g)), which indicates that MBP-Ce6 NSs could effec-tively decompose H2O2 to release O2. The generation of O2could relieve the hypoxia of bacterial biofilm-infected tissuesand may further enhance the efficacy of oxygen-dependentaPDT. Therefore, the singlet oxygen (1O2) production ofMBP-Ce6 NSs in hypoxic conditions with the presence ofH2O2 was measured and evaluated by using 2′,7′-dichlor-odihydrofluorescein diacetate (DCFH-DA) as the fluores-cent reactive oxygen species (ROS) indicator [52–54]. Asshown in Figure S7(a) (Supporting Information), the 1O2generation efficiency of MBP-Ce6 NSs is less effective thanfree Ce6 in normoxic condition due to the fluorescencequenching of MnO2 NSs. In contrast, MBP-Ce6 NSs canefficiently produce high-level 1O2 after laser irradiation inhypoxic conditions with the presence of H2O2 similar to thebiofilm microenvironment, while Ce6, H2O2, Ce6 + H2O2,and MBP-Ce6 NSs generate neglectable 1O2 (Figure S7(b),Supporting Information), suggesting that MBP-Ce6 NSs canserve as efficient agent during aPDT process in hypoxicconditions. These results mentioned above are consistentwith the elevated concentration of O2 shown in Figure 2(g).

The colloidal stability and cytotoxicity of MBP-Ce6 NSswere assessed before further biological applications. Afterbeing dispersed in various conditions (H2O, PBS, andDulbecco’s modified Eagle’s medium (DMEM)), MBP-Ce6 NSs showed no obvious aggregation (Figure S8(a),Supporting Information). Moreover, the change of theirhydrodynamic diameters within 24 h was neglectable(Figure S8(b), Supporting Information), indicating excellentcolloidal stability of MBP-Ce6 NSs. Then, MPB-Ce6 NSswith different concentrations (MnO2: 0, 20, 40, 80, 160, and320μg/mL; Ce6: 0, 16, 32, 64, 128, and 256μg/mL) werecultivated with human prostatic stromal myofibroblast(WPMY-1) cells for 24 h. Results show that MBP-Ce6NSs have little cytotoxicity even at high concentration(Figure S9, Supporting Information).

2.2. In Vitro Photodynamic Treatment of MRSA Biofilms.To investigate the Ce6 release and fluorescence imagingabilities of MBP-Ce6 NSs in bacterial biofilms in vitro,MRSA biofilms were used as a model. For comparison,WPMY-1 cells were adopted as the control to mimic thenormal environment condition. After being treated withMBP-Ce6 NSs (MnO2 NSs: 100μg/mL; Ce6: 80μg/mL)at different times, gradually increased red fluorescence(Ce6) in MRSA biofilms could be observed from the 3Dconfocal laser scanning microscopy (3D CLSM) imagesduring incubation (Figure 3(a)), while no distinct fluores-cence of Ce6 could be seen for WPMY-1 cells at the sameexperimental conditions (Figure 3(b)). The fluorescenceintensity of Ce6 incubated with MRSA biofilms is about

2.5 times higher than that of WPMY-1 cells at 12 h(Figure 3(c)), suggesting MBP-Ce6 NSs have specific biofilmmicroenvironment-responsive releasing ability. The sizedecrease of MBP-Ce6 NSs may improve the penetration ofCe6 in biofilms and further enhance their PDT effect.

The MR imaging ability of MBP-Ce6 NSs in bacterialbiofilm microenvironment was further evaluated. As shownin Figure 3(d), the T1-weighted MR signal of MBP-Ce6 NSsgradually increased in acidic solution (pH = 5:0) with theincrease of their concentration, while that in neutralcondition (pH = 7:4) showed a much weaker MR signal.Figure 3(e) indicates that the transverse relativity (r1) ofMBP-Ce6 NSs after being incubated in acidic buffer(pH = 5:0) for 24h is 10.99mM-1 s-1, which is about 3 timesof that in neutral buffer (3.47mM-1 s-1, pH = 7:4). Such sig-nificantly enhanced MR signal proves that MBP-Ce6 NSscould serve as pH-responsive MRI contrast agents.

Then, the efficiency of MBP-Ce6 NSs for the treatment ofMRSA biofilms in vitro was evaluated. Based on the 3DCLSM images shown in Figure 3(f), MRSA biofilms treatedby MBP-Ce6 NSs with both laser irradiation and H2O2 havethe lowest green fluorescence, indicating their excellent ther-apeutic effect. In contrast, MRSA biofilms in other groupswith different treatments (Ce6 + laser, H2O2 + laser, andMBP-Ce6 NSs + laser) showed much stronger fluorescence,suggesting limited antibiofilm effect. The in vitro aPDT effi-ciency of MBP-Ce6 NSs for MRSA biofilms was also studied.As shown in Figure 3(g), the colony-forming units (CFU)number of MRSA biofilms treated by MBP-Ce6 NSs(MnO2 NSs: 100μg/mL, Ce6: 80μg/mL) and H2O2 (50μM)with the laser irradiation decreased by 1.5 log (97%). In com-parison, the bacteria inactivation efficiency is less than 0.6 log(75%) for other groups (Ce6 + laser, Ce6 + H2O2 + laser, andMBP-Ce6 + laser). For the MBP-Ce6 + H2O2 and MBP +H2O2 + laser groups, no obvious antibacterial effect can beobserved. These results prove that the MBP-Ce6 NSs canachieve excellent antibiofilm performance by aPDT withthe presence of H2O2, which may be converted into O2 bythe MnO2 catalyzing and facilitate the generation of 1O2.To verify this assumption, the generation of 1O2 in MRSAbiofilms after different treatments was tested. As indicatedin Figure S10 (Supporting Information), MRSA biofilmstreated by MBP-Ce6 NSs with laser and H2O2 showedmuch stronger fluorescence than those treated by saline,Ce6, and MBP with laser and H2O2, suggesting a higherlevel of 1O2. SEM images (Figure 3(h)) reveal that theMRSA biofilms in MBP-Ce6 + laser + H2O2 group showdistinctly structure damage, which can also be observedfrom the images of MRSA biofilms stained by crystal violetwith decrease of relative biofilm biomass (Figure S11,Supporting Information).

2.3. Microenvironment-Responsive Dual-Mode Imaging ofMRSA Biofilm Infection. In order to evaluate the in vivoimaging capability of MBP-Ce6 NSs, MRSA biofilm-infected mice models were constructed by subcutaneouslyinjection of MRSA dispersions into the right thigh (biofilminfected) of each mouse (Figure 4(a)). In comparison, PBSdispersion was also used to treat the left thigh (control) of

6 Research

4000

WPMY-1MRSA

1 h

Calcein-AM

Ce6

Merged

Calcein-AM

pH 7.4

pH 5.0mM0

50

pH 7.4 r1 = 3.45 mM–1s–1

pH 5.0 r1 = 10.99 mM–1s–1

40

30

20

1/T 1

(s–1

)

100

0

9

MBP-Ce6+H2O2

MBP-Ce6+laserCe6+laser

MBP+H2O2+laser

MBP-Ce6+H2O2+laserCe6+H2O2+laser

8

7

0MnO2:Ce6:

10 50 1000 8

Concentration (𝜇g/mL)40 80

Log

CFU

/mL

0.8 1.6Mn concentration (mM)

2.4 3.2

3.2

Ce6

Merged

Saline

MBP

Ce6

MBP-Ce6

SalineH2O2 H2O2+laser

Laser : ––

–+

+–

++H2O2 :

4 h 8 h 12 h

1 h 4 h 8 h 12 h

3000

2000

1000Inte

nsity

(a.u

.)

0 1 4 8Time (h)

12

⁎⁎⁎⁎⁎⁎⁎

MBPH2 O2 H2O2+laser

MBP-Ce6H2O2 H2O2+laser

(a)

(b)

(f)

(d)

(e)

(g)

(h)

(c)

Figure 3: In vitro Ce6 releasing behavior of MBP-Ce6 NSs and photodynamic treatment of MRSA biofilms. (a) 3D CLSM images of MRSAbiofilms after being incubated with MBP-Ce6 (MnO2 NSs: 100μg/mL; Ce6: 80μg/mL) at different times. The scale bar is 200μm. (b) CLSMimages ofWPMY-1 cells after being treated with MBP-Ce6 (MnO2 NSs: 100μg/mL; Ce6: 80μg/mL) at different times. The green fluorescenceand the red fluorescence originate from Calcein-AM and Ce6, respectively. The scale bar is 200 μm. (c) The fluorescence intensity of Ce6 afterMBP-Ce6 NSs (MnO2 NSs: 100 μg/mL; Ce6: 80 μg/mL) incubated with MRSA biofilms andWPMY-1 cells at different times. (d) T1-weightedMR images of MBP-Ce6 NSs solutions with different concentrations and pH values in PBS. (e) The transverse relativities (r1) for MBP-Ce6NSs under different pH values in PBS. (f) 3D CLSM images of live bacteria (green fluorescence) insideMRSA biofilms after various treatmentsand stained with Calcein-AM. The scale bar is 200 μm. (g) The CFU number of bacteria inside MRSA biofilms after different treatments.∗∗p < 0:01 and ∗∗∗p < 0:001 (two-tailed Student’s t-test). (h) SEM images of MRSA biofilms treated with saline, MBP NSs (MnO2 NSs:100μg/mL), and MBP-Ce6 NSs (MnO2 NSs: 100 μg/mL; Ce6: 80 μg/mL) under different conditions. The scale bar is 2μm. Theconcentration of H2O2 is 50 μM. The laser irradiation was carried out by using the power density of 20mW/cm2 at 635 nm for 30min.Error bars indicate s.d. (n = 3).

7Research

PBS MRSA

MRSA biofilm infection–2 day

1 h(a)

(b)

(c)

(d)

(e)

(f) (h)

(g)

1 h 4 h 8 h 12 h

1 h

4 h

8 h

12 h

24 h

24 h High

Low

1

4

3

2

1

04 8 12

Time (h)

Aver

age fl

uore

scen

ce in

tens

ity(1

09 , p s–1

cm–2

sr–1

)

24

8 h 12 h 24 h High0 min

PBS MRSA

PBS

PBS

MRSA

MRSA

MRSA biofilms MRI

FLI

1, 4, 8, 12, and 24 h

30 min

120 minLow

0 daySubcutaneous injection

of MBP-Ce6 NSs

Imaging at different times

MBP-Ce6 NSs MBP-Ce6 NSs

MRSA biofilms MRI

FLI

MRSA

MRSA biofilm infection–2 day 0 day

Intravenous injectionof MBP-Ce6 NSs

Imaging at different times

MBP-Ce6 NSs

⁎⁎⁎

1

6000

4500

3000

1500

4 8 12Time (h)

T 1 sign

al in

tens

ity (a

.u.)

24

⁎⁎⁎

PBS MRSA PBS MRSA PBS MRSA PBS MRSA

Figure 4: In vivo fluorescence imaging (FLI) and magnetic resonance imaging (MRI) of MRSA biofilm infections by using MBP-Ce6 NSs.(a) Schematic illustration for the detection of MRSA biofilms by in situ injection of MBP-Ce6 NSs (50 μL; MnO2: 50 μg/mL; and Ce6:40μg/mL) both in the left thigh (normal tissue) and the right thigh (infected tissue). (b) Fluorescence images and (c) T1-weighted MRimages of the infected mice at different times. (d) Schematic illustration of the process for the detection of MRSA biofilms by i.v.injection of MBP-Ce6 NSs. (e) Fluorescence images and the (f) fluorescence intensity of MRSA-infected mice at different timespostinjection. (g) T1-weighted MR images shown in cross-section and (h) the intensity of the T1-weighted MR signal of theMRSA biofilm-infected tissues at different times postinjection. White dashed circles in (e) and (g) denote MRSA biofilm-infected tissues.∗∗∗p < 0:001 (two-tailed Student’s t-test). All data are presented as means ± s:d: (n = 3).

8 Research

the same mice. Two days later, MRSA biofilm infections withsignificant abscesses (Figure S12(a), Supporting Information)appeared and could be confirmed by wound blotting(Figure S12(b), Supporting Information) [55, 56]. Thein vivo imaging of MRSA biofilms was first studied by insitu injection of MBP-Ce6 NSs.

As illustrated in Figure 4(b) and (c), and Figure S13(Supporting Information), the biofilm-infected tissues(right thigh) showed significantly enhanced fluorescenceand T1-weighted MR signals after local injection of MBP-Ce6 NSs compared with normal tissues (left thigh), whichsuggests that detection of the MRSA biofilm infectionscould be achieved by MBP-Ce6 NSs. To further examinethe in vivo FL/MR dual-mode imaging ability, MBP-Ce6NSs were used to detect MRSA biofilm infections byintravenous (i.v.) injection (Figure 4(d)). Figure 4(e) and (f)show that the fluorescence of the infected tissues in micetreated by i.v. injection of MBP-Ce6 NSs increasedgradually and peaked at the 8th hour postinjection. After24 h postinjection, the fluorescence signal in biofilm-infected tissues was much stronger than other organs(Figure S14, Supporting Information), indicating specificaccumulation in infected tissues due to the EPR effect andpH-responsive decomposition of MBP-Ce6 NSs [29, 42].Meanwhile, the T1-weighted MR signal of the MRSAbiofilm infections also increased obviously during the first8 h and gradually decreased afterwards (Figure 4(g) and (h)),similar to the change of fluorescence signal. As Figure S15(Supporting Information) shows, the T1-weighted MR signalin the kidneys was much stronger than that in other organsduring 24h postinjection and gradually became weaker afterthe 8th hour postinjection, suggesting that the Mn2+ ionsreleased from MBP-Ce6 NSs could be quickly excretedthrough the kidney [45]. In addition, we also found thatthe amount of Mn in biofilm-infected tissues is much higherthan that in the other organs, indicating the specificaccumulation of MBP-Ce6 NSs in MRSA biofilm-infectedtissues (Figure S16, Supporting Information). Furthermore,the content of Mn in the organs quickly decreased tolow level within 48 h, which demonstrates that MBP-Ce6NSs could be quickly excreted from the body afterdecomposition within relatively short periods [42].

2.4. Relieving Hypoxia for Enhanced PDT of MRSA BiofilmInfection. During the development of bacterial biofilms, O2is consumed by both the bacteria inside the biofilms andthe surrounding inflammatory cells, while the diffusion ofO2 into the biofilms is hampered by the existence of the denseEPS, which together cause the hypoxia in bacterial biofilmmicroenvironment [47, 57]. It is known that PDT is anoxygen-dependent process and the therapeutic efficacy canbe greatly restricted by the hypoxic conditions. Hence, thehypoxia microenvironment in biofilms is an intrinsic barrierfor aPDT treatment. As reported, excess H2O2 usually existsin biofilm-infected tissues [4, 28], in principle, which couldbe converted to O2 with the catalysis of MnO2. Herein, thepotential of MBP-Ce6 NSs to generate O2 and relieve thehypoxia of biofilm was investigated in a mice model. Afteri.v. injection with PBS, Ce6, MBP NSs, and MBP-Ce6 NSs,

the biofilm-infected tissues of mice were harvested at 8thhour posttreatment and stained with fluorescent dye-labeled HIF-1α antibody [58].

As shown in Figure 5(a), CLSM images indicate that theMRSA biofilm-infected tissues treated with PBS or Ce6showed a larger hypoxia-positive area (green), while thosetreated with MBP NSs or MBP-Ce6 NSs exhibited a limitedhypoxia-positive area. The semiquantification analysisresults (Figure 5(b)) further confirm that the treatment byMBP NSs or MBP-Ce6 NSs could effectively relieve the hyp-oxia conditions inside MRSA biofilm-infected tissues. Thein vivo therapeutic efficacy of aPDT by using MBP-Ce6 NSswas further studied. The mice with MRSA biofilm infectionswere divided into different groups according to differenttreatments: laser irradiation (group 1, laser), MBP NSswith laser irradiation (group 2, MBP + L), Ce6 with laserirradiation (group 3, Ce6 + L), MBP-Ce6 NSs (group 4,MBP-Ce6), and MBP-Ce6 NSs with laser irradiation(group 5, MBP-Ce6 + L). After i.v. injection of the agents(dose of MnO2 = 5mg/kg; dose of Ce6 = 4mg/kg) for 8 h,laser irradiation under 635nm laser (20mW/cm2) was per-formed in groups 1, 2, 3, and 5 for 30min. At therapeuticday 8, themice in group 5 (MBP-Ce6 NSs + laser) showed sig-nificantly reduction of the infected area compared with theother four groups (Figure 5(c)). The abscesses and woundbeds of the mice in group 5 disappeared, while the mice withother treatments (groups 1, 2, 3, and 4) still had a large areaof infection (Figure 5(d)). In addition, the infected tissueswere harvested and the number of viable bacteria insidebiofilm-infected tissues was evaluated at 8th day posttreat-ment. Figure 5(e) shows that the bacterial inactivation effi-ciency of group 5 (MBP-Ce6 NSs + laser) is about 2.5 log(99.7%), much higher than those in group 1 (0%), group 2(0%), group 3 (80.1%), and group 4 (0%). As shown in thehematoxylin-eosin (H&E) staining images (Figure 5(f);Figure S17, Supporting Information), severe inflammationwith massive inflammatory cells infiltration still existed ininfected tissues of the mice in groups 1-4 (indicated by bluearrows). In contrast, the infected tissues in group 5 (MBP-Ce6 NSs + laser) showed obvious proliferation of fibroblastcells, neovascularization (indicated by green cycles), and alittle granulomatous inflammation (indicated by red cycles).Masson’s trichrome staining images further indicate theformation of intact subcutaneous tissues and the appearanceof more collagen fibers (blue) in group 5 than the othergroups (groups 1, 2, 3, and 4), suggesting better recovery ofthe infected tissues. These results demonstrated that MBP-Ce6 NSs have excellent aPDT efficacy for MRSA biofilminfections, which should be ascribed to the hypoxia-relief inthe biofilm microenvironment.

Furthermore, long-term toxicity ofMBP-Ce6 NSs tomicewas also studied. As shown in Figure S18(a) (SupportingInformation), H&E staining images of major organs fromthe mice after i.v. injection of MBP-Ce6 NSs show nonoticeable damage or inflammatory lesion. Figure S18(b)(Supporting Information) indicates that the body weights ofthe mice treated by MBP-Ce6 NSs steadily increased andhad no obvious difference from the mice treated by PBS.The hematology assay results (Figures S19 (c-j), Supporting

9Research

PBS

Nucleus

Hypoxia

Overlay

Day 0

Lase

rM

BP+l

aser

Ce6

+las

erM

BP-C

e6M

BP-C

e6+l

aser

Day 8

MBP MBP-Ce650

40

30

20

10

Hyp

oxia

-pos

itive

area

(%)

0

(a)

(d)

(f)

(b)

(c)

(e)

PBS

35

28

21

14

Infe

cted

area

(mm

2 )

7

00

12

10

8

6Log

CFU

(per

tiss

ue)

2 4 6Time (day)

8 10

LaserMBP+laserCe6+laserMBP-Ce6MBP-Ce6+laser

Laser

H&E

Masson

MBP+laser Ce6+laser MBP-Ce6 MBP-Ce6+laser

Lase

r

MBP

+las

er

Ce6

+las

er

MBP

-Ce6

MBP

-Ce6

+las

er

Ce6 MBP MBP-Ce6

Ce6

⁎⁎⁎⁎⁎⁎

⁎⁎⁎⁎

⁎⁎

⁎⁎⁎⁎⁎⁎

⁎⁎⁎⁎⁎⁎

Day 0 Day 8

50

40

30

20

10

Hyp

oxia

-pos

itive

area

(%)

0

(a)

(d)

(b)

(c)

(e)

PBS

35

28

21

14

Infe

cted

area

(mm

2 )

7

00

12

10

8

6Log

CFU

(per

tiss

ue)

2 4 6Time (day)

8

LaserMBP+laserCe6+laserMBP-Ce6MBP-Ce6+laser

Laser MBP+laser Ce6+laser MBP-Ce6 MBP-Ce6+laser

Lase

r

MBP

+las

er

Ce6

+las

er

MBP

-Ce6

MBP

-Ce6

+las

er

Ce6 MBP MBP-Ce

⁎⁎⁎

⁎⁎⁎⁎

⁎⁎

⁎⁎⁎⁎⁎⁎

⁎⁎⁎⁎⁎⁎

Figure 5: Photodynamic treatment of MRSA biofilm infections in vivo. (a) CLSM immunofluorescence images of the infected tissueslices from the mice after i.v. injection with PBS, Ce6, MBP NSs, and MBP-Ce6 NSs for 8 h, respectively. The hypoxic conditionwas indicated by staining the slices with HIF-1α antibody (green) and 4′,6-diamidino-2-phenylindole (DAPI, blue). The scale bar is200μm. (b) Semiquantification of the hypoxia-positive area in biofilm-infected tissues after various treatments. (c) Biofilm-infected tissueareas of mice after different treatments at different times. (d) Photographs of the biofilm-infected tissues from the mice with varioustreatments at different times. (e) CFU of MRSA in biofilm-infected tissues at 8th day posttreatment. (f) Histological analysis of thebiofilm-infected tissues by H&E and Masson’s trichrome staining. Blue arrows denote the inflammatory cell infiltration. Green circleindicates the proliferation of fibroblast and neovascularization. Red circle indicates granulomatous inflammation. The scale bar is 200 μm.∗∗∗p < 0:001 (two-tailed Student’s t-test). All data are presented as means ± s:d: (n = 5).

10 Research

Information) further show that all serum biochemicalparameters and blood routine examination parameters ofthe mice treated by MBP-Ce6 NSs are similar with thecontrol group at 28 d postinjection, further indicating lowtoxicity of MBP-Ce6 NSs at the dose used in this study [59].

3. Conclusion

In summary, we designed multifunctional nanotheranostics(MBP-Ce6 NSs) for biofilm microenvironment-responsiveimaging and treating of MRSA biofilm infections. MBP-Ce6NSs could be decomposed in the acidic microenvironmentin biofilm-infected tissues and subsequently release Ce6 andMn2+ ions. The FL signal from Ce6 and MR signal fromMn2+ could be simultaneously activated during the decom-position of MBP-Ce6 NSs. Experimental results show thatMBP-Ce6 NSs can achieve FL/MR dual-mode imaging forspecific and sensitive detection of MRSA biofilm infections.On the other hand, the hypoxic condition in biofilm micro-environment could be relieved by MnO2-triggered decompo-sition of endogenous H2O2 in biofilm-infected tissues, whilehypoxia-associated resistance of biofilm toward aPDT couldbe overcome. After being treated by MBP-Ce6 NSs withlaser irradiation, the number of viable MRSA from biofilm-infected tissues in mice could be reduced by 2.5 log(99.7%), which is much lower than the control groups, andthe recovery of the infected tissues could be promoted. Fur-thermore, the MBP-Ce6 NSs showed no appreciable toxicityin the in vivo study, indicating their potential use in biomed-icine. Therefore, this study provides a promising theranosticnanoplatform with potential use for specific detection andtreatment of bacterial biofilm infections.

4. Materials and Methods

4.1. Materials. Manganese chloride tetrahydrate(MnCl2•4H2O), tetramethylammonium hydroxide solution(TMA, 1.0M), bovine serum protein (BSA), and 2′,7′-dichloroflourescin diacetate (DCFH-DA) were purchasedfrom Sigma-Aldrich. NH2-PEG3000 and chlorin e6 (Ce6)were purchased from JenKem Technology and Frontier,respectively. The hydrogen peroxide (H2O2, 30wt.%) solu-tion was purchased from Sinopharm Chemical Reagent Co.,Ltd. Dulbecco’s modified Eagle’s medium (DMEM) and fetalbovine serum (FBS) were bought from KeyGen BioTech andGibco, respectively. HIF-1α antibody (mouse monoclonalantibody) and FITC-labeled goat anti-mouse IgG werepurchased from Beyotime Biotechnology. All solutions wereprepared by using ultrapure water (18.2MΩ).

4.2. Characterization. TEM images were obtained by using anHT7700 transmission electron microscope (Hitachi, Japan).Anatomic force microscope (AFM, Nanoscope IIIa, Bruker,USA) was used to measure the thickness of the nanosheets.X-ray diffraction (XRD) patterns were recorded on a D8ADVANCE X-ray diffractometer (Bruker, Germany) withCu Kα radiation. The ultraviolet-visible-infrared (UV-Vis-NIR) absorption spectra were recorded on a UV-3600spectrophotometer (Shimadzu, Japan). Fluorescence spectra

(FL) were recorded on a Shimadzu RF-5301 PC fluorescencespectrophotometer. The concentration of O2 in solutions wasmeasured by using a dissolved oxygen detection probe(HACH, China). Photodynamic therapy (PDT) was per-formed by using a 635nm continuous-wave semiconductorlaser (BWT Beijing, China). The power density of laserirradiation was measured by using a digital power meter(PM100D, Thorlabs, USA). Dynamic light scattering (DLS)and zeta potential were performed on a ZetaPALS PotentialAnalyzer (Brookhaven Instruments, USA).

4.3. Synthesis of MnO2 NSs. MnO2 nanosheets (MnO2 NSs)were prepared based on previous study. Twelve-milliliterTMA solution (0.1M) and 2mL H2O2 (30wt.%) were mixedwith 6mL ultrapure water as solution A. MnCl2•4H2O (0.6 g)was dissolved in 10mL H2O as solution B. Then, solution Awas quickly mixed with solution B and the color of themixture became dark brown indicating that the Mn2+ wasoxidized to Mn4+. After stirring overnight, the aggregatedMnO2 NSs were washed with ethanol and water for threetimes, respectively, and were then dispersed in 30mLultrapure water. To prepare single-layer MnO2 NSs, theas-prepared MnO2 aggregations were exfoliated by ultrasoni-cation (Yimaneili, 950W, 25 kHz) for 10h in an ice-bath. Themixtures were centrifuged at 12000 rpm for 45min to sepa-rate large aggregations and were further centrifuged at18000 rpm for 90min to obtain the single-layer MnO2 NSs.This step was repeated for three times. The MnO2 NSs weredispersed in ultrapure water and then stored at 4°C. Theconcentration of MnO2 NSs was determined by using aninductively coupled plasma atomic emission spectrome-ter(ICP-AES, Optima 5300DV, PerkinElmer).

4.4. Preparation of MnO2-BSA/PEG-Ce6 NSs. MnO2 NSs(5mg) and BSA (100mg) were dispersed in 15mL waterand stirred for 12h. Then, the mixture was centrifuged at16000 rpm for 45min to remove any large aggregations.The MnO2-BSA nanosheets were dispersed in 15mL waterand mixed with 50mg PEG3000-NH2 by stirring for 12 h.The mixture was centrifuged at 16000 rpm for 45min andwashed with water for three times to get MnO2-BSA/PEGnanosheets (MBP NSs). For Ce6 loading, MBP NSs andCe6 aqueous dispersions were mixed in an orbital shakerfor 12 h and then centrifuged at 16000 rpm for 45min toobtain MnO2-BSA/PEG-Ce6 nanosheets (MBP-Ce6 NSs).MBP-Ce6 NSs aqueous dispersions were stored at 4°C.

The loading efficiency and encapsulation efficiency ofCe6 by MBP NSs were determined according to the followingequations

Ce6 loading efficiency %ð Þ =W1/W2 × 100% ð1Þ

Ce6 encapsulation efficiency %ð Þ = Wt −Wsð Þ/ Wtð Þ × 100%ð2Þ

where W1 is the weight of Ce6 loaded in MBP-Ce6 NSs, W2is the total weight of MBP-Ce6 NSs, Wt represents the totalweight of Ce6 initially added, and Ws represents the weightof Ce6 remaining in the supernatant.

11Research

4.5. Degradation and Drug Release Study of MBP-Ce6 NSs. Tostudy the degradation process, MBP-Ce6 NSs were dispersedin PBS solutions with different pH values (5.0, 6.0, and 7.4).At given time points, the MBP-Ce6 NS PBS dispersions werecollected and quantified by using UV-Vis absorption spec-tra. For Ce6 release studies, the MBP-Ce6 NSs were dis-persed in PBS solutions with different conditions (pH7.4in PBS, pH5.0 in PBS, pH7.4 in PBS with 50μM H2O2,and pH5.0 in PBS with 50μM H2O2). The amount ofreleased Ce6 was determined by using UV-Vis absorptionspectra, and the fluorescence recovery was monitored byusing fluorescence spectra.

4.6. Detection of Singlet Oxygen during PDT. 2′, 7′-Dichlor-ofluorescein diacetate (DCFH-DA) was used to detect thegeneration of singlet oxygen. In order to detect the photody-namic effect of MBP-Ce6 NSs in hypoxia conditions, O2 wereremoved from the PBS solutions by bubbling with N2 for30min. Then, DCFH-DA (10mg/mL, 5μL) and Ce6(1mg/mL, 10μL) or MBP-Ce6 (containing 10μg Ce6) wereadded into 3mL PBS solutions either with or without H2O2(50μM), after being irradiated with a 635nm laser at thepower density of 20mW/cm2. The fluorescence spectra ofthese solutions were collected at different times.

4.7. Cytotoxicity of MnO2 NSs and MBP-Ce6 NSs. A LDH-Cytotoxicity Colorimetric Assay Kit (BioVision) was usedto evaluate the cytotoxicity of MnO2 NSs and MBP-Ce6NSs. Human normal prostatic myofibroblast (WPMY-1)cells were cultured in DMEM (Dulbecco’s modified Eagle’smedium, KeyGen, BioTech) with 10% fetal bovine serum(FBS, Gibco) medium at 37°C under 5% CO2. WPYM-1 cellswere added in 96-well plates with the concentration of 104

cell per well. After being cultured for 24 h, the supernatantmedium was removed and MnO2 NS and MBP NS disper-sions (200μL, suspended in DMEM (without FBS)) contain-ing different concentrations of MnO2 (0, 20, 40, 80, 160, and320μg/mL) were added into the 96-well plate. Cells culturedwithout any nanosheets were set as low control, and thosecultured with 1% Triton X-100 were used as high control.After 24 h, 100μL of supernatant in each well was transferredinto another 96-well plate, and 100μL reaction solution wasadded into the corresponding well. After 30min, a micro-plate reader (PowerWave XS2, BioTek) was used to measurethe optical density at 495nm (OD495). The viability of cellswas calculated according to the following formula

Cell viability = 100% − AT – ALð Þ/ AH – ALð Þ × 100% ð3Þ

where AT represents the OD495 of test sample, AL representsthe OD495 of low control, and AH represents the OD495 ofhigh control.

4.8. Bacterial Culture. Methicillin-resistant Staphylococcusaureus (MRSA, ATCC43300) was used in this study. Beforebeing used, a single colony of MRSA was transferred into5mL Luria-Bertani (LB) culture and cultured for 10 h. Thebacterial dispersion was washed with saline and quantitatedby a microplate reader. To obtain mature MRSA biofilms,

MRSA dispersion in LB culture (containing 1% glucose,107CFU/mL) was cultured at 37°C for 48 h.

4.9. Ce6 Release in MRSA Biofilms. For pH-responsive Ce6release study, MRSA biofilms were cultured in a confocalpetri dish. The MBP NSs (MnO2: 100μg/mL; Ce6: 80μg/mL)were cultured with MRSA biofilms at 37°C at different times(1, 4, 8, and 12h). After being washed with saline for threetimes, MRSA biofilms were stained with Calcein-AM(KeyGen BioTech) for 30min and then imaged by anOlympus IX81 confocal laser scanning microscope. The3D images were obtained by using the FV10-ASW Viewer.

4.10. In Vitro Photodynamic Treatment of MRSA Biofilms. Totreat the MRSA biofilms by PDT, MBP NSs (MnO2: 10, 50,and 100μg/mL) and MBP-Ce6 NSs (MnO2: 10, 50, and100μg/mL and Ce6: 8, 40, and 80μg/mL) were added intoMRSA biofilms with or without H2O2 (50μM). After beingcultured for 12 h, the MRSA biofilms in the laser treatmentgroups were irradiated with 635 nm laser (20mW/cm2).MRSA biofilms after different treatments were harvestedby ultrasonication. The standard plate count method wasused to determine the number of viable bacteria insideMRSA biofilms.

For fluorescence imaging, the treated MRSA biofilmswere stained with Calcein-AM for 30min and imaged byusing a confocal laser scanning microscope (OlympusIX81). For crystal violet staining, MRSA biofilms were fixedwith formalin for 10min and dyed with crystal violet solution(0.02%) for 1 h. Images were captured by using a motorizedfluorescence microscope (Olympus IX71). For SEM imaging,MRSA biofilms were grown on the surface of an indium tinoxide (ITO) glass. After treatment, MRSA biofilms were fixedin 2.5% glutaraldehyde solution for 30min, dehydrated withseries of ethanol solutions with gradient concentrations(15%, 30%, 50%, 75%, and 100%) for 15min, respectively.After sputtering coated with gold, the morphology character-ization of bacteria was performed on a Hitachi S4800 SEM.

4.11. MRSA Biofilm-Infected Animal Models. All animal pro-cedures were performed in accordance with the Guidelinesfor Care and Use of Laboratory Animals, and experimentswere approved by the Animal Ethics Committee of NanjingUniversity. Female Balb/c mice (20 g, 6 weeks old) were pur-chased from Nanjing Junke Biological Engineering Co., Ltdand used in this study. To develop the MRSA biofilm-infected mice model, MRSA suspensions in PBS (50μL,109CFU/mL) were subcutaneously injected into the miceskin. After being infected for 2 days, subcutaneous abscessesappeared in the treated mice, indicating the formation ofMRSA biofilms.

4.12. FL and MR Imaging of MRSA Biofilm-Infected Mice.Before the FL and MR imaging, MBP-Ce6 NSs dispersionswere intravenously (i.v.) injected into MRSA-infected mice(dose of MnO2 = 5mg/kg, dose of Ce6 = 4mg/kg). Fluores-cence images of mice at different times (1 h, 4 h, 8 h, 12 h,and 24 h) postinjection were captured on an IVIS LuminaK Series III system (Perkin Elmer). MR imaging was con-ducted on a MRI scanner system (BioSpec 7T/20 USR,

12 Research

Bruker) at different times (1 h, 4 h, 8 h, 12 h, and 24 h) postin-jection of MBP-Ce6 NSs. The MRI images were obtained byusing the Sante MRI Viewer.

4.13. Photodynamic Therapy of MRSA Biofilm-Infected Mice.Thirty mice with MRSA biofilm infections were randomlydivided into five groups: (1) saline + laser, (2) MBP NSs +laser, (3) Ce6 + laser, (4) MBP-Ce6 NSs, and (5) MBP-Ce6NSs + laser. The mice in groups 1-3 were i.v. injected withsaline (200μL), MBP NS dispersion in PBS (200μL; MnO2:5mg/kg), Ce6 dispersion in PBS (200μL; dose of Ce6:4mg/kg), respectively. The mice in groups 4 and 5 were i.v.injected with MBP-Ce6 NSs dispersions (200μL; MnO2:5mg/kg, Ce6: 4mg/kg). At 8th hour postinjection, the micein the laser treatment groups (groups 1, 2, 3, and 5) were irra-diated with 635nm laser at the power density of 20mW/cm2

for 30min. The size of the infected tissues was measured byusing caliper every 2 days. The area of the infected tissueswas calculated according to the formula: area = ðwidth/2 ×length/2Þ × π. After the treatment for 8 days, the infected tis-sues of the mice were harvested and ultrasonicated for 10minto disperse the bacteria. Then, the CFU numbers of the bac-teria in the infected tissues were determined by the platecount method.

For histological analysis, the infected tissues of micewere harvested and fixed with 4% polyoxymethylene andthen processed by standard protocols for hematoxylin andeosin (H&E) staining and Masson’s trichrome staining,respectively.

4.14. Immunofluorescence Staining of MRSA Biofilm-InfectedTissues. MRSA biofilm-infected mice were i.v. injected withsaline (200μL), Ce6 (200μL, 4mg/kg), MBP NSs (200μL,MnO2: 5mg/kg), and MBP-Ce6 NSs (200μL, MnO2:5mg/kg, Ce6: 4mg/kg), respectively. At 8th hour postinjec-tion, the infected tissues of mice were harvested, fixed with4% polyoxymethylene, embedded in paraffin, and sectionedinto slices. Triton X-100 solution (0.5%) was used for celllysis. BSA was used to prevent nonspecific adsorption. Thetissue slices were incubated with HIF-1α mouse monoclonalantibody at 4°C overnight. Then, the FITC-labeled goat anti-mouse secondary antibody was added at 37°C for 2 h. Cellnuclei were stained with 4′,6-diamidino-2-phenylindole(DAPI). Fluorescence images of the stained tissue slices wereobtained by using the confocal laser scanning microscope(Olympus IX81).

4.15. Toxicity Study of MBP-Ce6 NSs In Vivo. Twelve Balb/cmice were randomly divided into two groups (6 mice pergroup) and were i.v. injected with 200μL PBS (group a)and 200μL MBP-Ce6 NSs (group b; MnO2: 10mg/kg, Ce6:8mg/kg), respectively. At 28th day postinjection, one mousewas sacrificed to afford the major organs (heart, liver, spleen,lung, and kidney) for H&E staining. The blood of theother five mice was collected for hematology assay. Forthe blood chemistry analysis, the levels of alkaline phos-phatase (ALP), alanine aminotransferase (ALT), aspartateaminotransferase (AST), and urea nitrogen (BUN) weremeasured. For the complete blood panel test, the levels

of white blood cells (WBC), red blood cells (RBC), plate-lets (PLT), hemoglobin (Hgb), hematocrit (HCT), meancorpuscular volume (MCV), mean corpuscular hemoglobinconcentration (MCHC), and mean corpuscular hemoglobin(MCH) were measured.

Data Availability

All data needed to evaluate the conclusions in the paper arepresent in the paper. Additional data related to this papermay be requested from the authors.

Conflicts of Interest

The authors declare no competing financial interest.

Authors’ Contributions

L. Wang and L. Yuwen conceived the idea and supervised thestudy. W. Xiu carried out the experiments, analyzed theresults, and wrote the manuscript. S. Gan helped with thein vivo experiments. Q. Wen, Q. Qiu, and S. Dai contributedto the material characterization and data analysis. H. Dong,Q. Li, and Y. Mou analyzed the H&E-stained slices. L. Wengand Z. Teng contributed to the scientific discussion of themanuscript. All authors discussed the results and partici-pated in revising the paper.

Acknowledgments

This work was financially supported by the NationalKey Research and Development Program of China(2017YFA0205302), Natural Science Foundation of JiangsuProvince (BK20191382), the Priority Academic ProgramDevelopment of Jiangsu Higher Education Institutions(PAPD, YX030003), the Jiangsu Provincial Key Researchand Development Program (BE2018732), and the NaturalScience Key Fund for Colleges and Universities in JiangsuProvince (17KJA430011).

Supplementary Materials

Figure S1: TEM image of the as-prepared MnO2 NSs aggre-gations. Figure S2: large-scale TEM images and size statisticsof MnO2 NSs, MBP NSs, and MBP-Ce6 NSs. Figure S3:selected area electron diffraction pattern and elemental map-ping images of MnO2 NSs. Figure S4: large-scale AFM imagesand thickness statistics of MnO2 NSs, MBP NSs, and MBP-Ce6 NSs. Figure S5: hydrodynamic diameter and zeta poten-tial of MnO2 NSs, MBP NSs, and MBP-Ce6 NSs. Figure S6:fluorescence recovery of MBP-Ce6 NSs at different condi-tions. Figure S7: the generation of singlet oxygen by MBP-Ce6 NSs. Figure S8: the colloidal stability characterizationof MBP-Ce6 NSs. Figure S9: the cytotoxicity of MBP-Ce6NSs. Figure S10: the ROS in MRSA biofilms treated byMBP-Ce6 NSs and H2O2. Figure S11: the crystal violet stain-ing images of MRSA biofilms after photodynamic treatment.Figure S12: photographs of MRSA biofilm-infected tissuesand nitrocellulose membrane contacted with the infectedissues and followed by ruthenium red staining. Figure S13:

13Research

the quantitative data of the FL and MRI imaging for thebiofilm-infected mice. Figure S14: fluorescence images ofthe major organs and MRSA biofilm-infected tissues fromthe mice treated by MBP-Ce6 NSs. Figure S15: T1-weightedMR images of the mice in longitudinal section after i.v. injec-tion of MBP-Ce6 NSs. Figure S16: biodistribution of Mn inmajor organs and MRSA biofilm-infected tissues from themice treated by MBP-Ce6 NSs. Figure S17: analysis ofH&E-stained slices of MRSA-infected tissues from mice.Figure S18: the in vivo toxicity evaluation of MBP-Ce6 NSs.(Supplementary Materials)

References

[1] L. Hall-Stoodley, J. W. Costerton, and P. Stoodley, “Bacterialbiofilms: from the natural environment to infectious diseases,”Nature Reviews Microbiology, vol. 2, no. 2, pp. 95–108, 2004.

[2] Z. Chen, H. Ji, C. Liu, W. Bing, Z. Wang, and X. Qu, “A mul-tinuclear metal complex based DNase-mimetic artificialenzyme: matrix cleavage for combating bacterial biofilms,”Angewandte Chemie International Edition, vol. 55, no. 36,pp. 10732–10736, 2016.

[3] D. Davies, “Understanding biofilm resistance to antibacterialagents,” Nature Reviews Drug Discovery, vol. 2, no. 2,pp. 114–122, 2003.

[4] Q. Deng, P. Sun, L. Zhang et al., “Porphyrin MOF dots–based,function-adaptive nanoplatform for enhanced penetration andphotodynamic eradication of bacterial biofilms,” AdvancedFunctional Materials, vol. 29, no. 30, article 1903018, 2019.

[5] K. Lewis, “Persister cells, dormancy and infectious disease,”Nature Reviews Microbiology, vol. 5, no. 1, pp. 48–56, 2007.

[6] H. C. Flemming, J. Wingender, U. Szewzyk, P. Steinberg, S. A.Rice, and S. Kjelleberg, “Biofilms: an emergent form of bacte-rial life,” Nature Reviews Microbiology, vol. 14, no. 9,pp. 563–575, 2016.

[7] H.Wu, C. Moser, H. Z.Wang, N. Høiby, and Z. J. Song, “Strat-egies for combating bacterial biofilm infections,” InternationalJournal of Oral Science, vol. 7, no. 1, pp. 1–7, 2015.

[8] B. P. Conlon, E. S. Nakayasu, L. E. Fleck et al., “Activated ClpPkills persisters and eradicates a chronic biofilm infection,”Nature, vol. 503, no. 7476, pp. 365–370, 2013.

[9] A. Penesyan, M. Gillings, and I. Paulsen, “Antibiotic discovery:combatting bacterial resistance in cells and in biofilm commu-nities,” Molecules, vol. 20, no. 4, pp. 5286–5298, 2015.

[10] G. Fang, W. Li, X. Shen et al., “Differential Pd-nanocrystalfacets demonstrate distinct antibacterial activity againstGram-positive and Gram-negative bacteria,” Nature Commu-nications, vol. 9, no. 1, article 129, 2018.

[11] X. Ning, W. Seo, S. Lee et al., “PET imaging of bacterial infec-tions with fluorine-18-labeled maltohexaose,” AngewandteChemie International Edition, vol. 53, no. 51, pp. 14096–14101, 2014.

[12] D. K. Kang, M. M. Ali, K. Zhang et al., “Rapid detection ofsingle bacteria in unprocessed blood using integrated compre-hensive droplet digital detection,” Nature Communications,vol. 5, no. 1, article 5427, 2014.

[13] L. Hall-Stoodley, F. Z. Hu, A. Gieseke et al., “Direct detectionof bacterial biofilms on the middle-ear mucosa of childrenwith chronic otitis media,” JAMA, vol. 296, no. 2, pp. 202–211, 2006.

[14] M. K. S. Larsen, T. R. Thomsen, C. Moser, N. Høiby, and P. H.Nielsen, “Use of cultivation-dependent and -independenttechniques to assess contamination of central venous cathe-ters: a pilot study,” BMC Clinical Pathology, vol. 8, no. 1,p. 10, 2008.

[15] J. Tang, B. Chu, J. Wang et al., “Multifunctional nanoagents forultrasensitive imaging and photoactive killing of Gram-negative and Gram-positive bacteria,” Nature Communica-tions, vol. 10, no. 1, article 4057, 2019.

[16] Z. Zhao, R. Yan, X. Yi et al., “Bacteria-activated theranosticnanoprobes against methicillin-resistant Staphylococcusaureus infection,” ACS Nano, vol. 11, no. 5, pp. 4428–4438,2017.

[17] X. Ning, S. Lee, Z. Wang et al., “Maltodextrin-based imagingprobes detect bacteria in vivowith high sensitivity and specific-ity,” Nature Materials, vol. 10, no. 8, pp. 602–607, 2011.

[18] M. van Oosten, T. Schäfer, J. A. C. Gazendam et al., “Real-timein vivo imaging of invasive- and biomaterial-associated bacte-rial infections using fluorescently labelled vancomycin,”Nature Communications, vol. 4, no. 1, article 2584, 2013.

[19] K. E. DeMeester, H. Liang, M. R. Jensen et al., “Synthesis offunctionalized N-acetyl muramic acids to probe bacterial cellwall recycling and biosynthesis,” Journal of the AmericanChemical Society, vol. 140, no. 30, pp. 9458–9465, 2018.

[20] Z. Zhao, H. Fan, G. Zhou et al., “Activatable fluorescence/MRIbimodal platform for tumor cell imaging via MnO2 nano-sheet–aptamer nanoprobe,” Journal of the American ChemicalSociety, vol. 136, no. 32, pp. 11220–11223, 2014.

[21] S. Zhu, R. Tian, A. L. Antaris, X. Chen, and H. Dai, “Near-infrared-II molecular dyes for cancer imaging and surgery,”Advanced Materials, vol. 31, no. 24, article 1900321, 2019.

[22] E. Terreno, D. D. Castelli, A. Viale, and S. Aime, “Challengesfor molecular magnetic resonance imaging,” ChemicalReviews, vol. 110, no. 5, pp. 3019–3042, 2010.

[23] X. Pang, X. Liu, Y. Cheng et al., “Sono-immunotherapeuticnanocapturer to combat multidrug-resistant bacterial infec-tions,” Advanced Materials, vol. 31, no. 35, article 1902530,2019.

[24] A. Gupta, R. Das, G. Yesilbag Tonga, T. Mizuhara, and V. M.Rotello, “Charge-switchable nanozymes for bioorthogonalimaging of biofilm-associated infections,” ACS Nano, vol. 12,no. 1, pp. 89–94, 2018.

[25] T. F. Moriarty, J. S. Elborn, and M. M. Tunney, “Effect of pHon the antimicrobial susceptibility of planktonic and biofilm-grown clinical Pseudomonas aeruginosa isolates,” British Jour-nal of Biomedical Science, vol. 64, no. 3, pp. 101–104, 2007.

[26] P. O. Jensen, M. Kolpen, K. N. Kragh, and M. Kuhl, “Microen-vironmental characteristics and physiology of biofilms inchronic infections of CF patients are strongly affected by thehost immune response,” APMIS, vol. 125, no. 4, pp. 276–288,2017.

[27] Y. Liu, Z. Ren, G. Hwang, and H. Koo, “Therapeutic strategiestargeting cariogenic biofilm microenvironment,” Advances inDental Research, vol. 29, no. 1, pp. 86–92, 2018.

[28] D. Mao, F. Hu, Kenry et al., “Metal-organic-framework-assisted in vivo bacterial metabolic labeling and precise anti-bacterial therapy,” Advanced Materials, vol. 30, no. 18, article1706831, 2018.

[29] D. S. Benoit and H. Koo, “Targeted, triggered drug delivery totumor and biofilm microenvironments,” Nanomedicine,vol. 11, no. 8, pp. 873–879, 2016.

14 Research

[30] D. S. W. Benoit, K. R. Sims Jr., and D. Fraser, “Nanoparticlesfor oral biofilm treatments,” ACS Nano, vol. 13, no. 5,pp. 4869–4875, 2019.

[31] Z. Xu, Z. Qiu, Q. Liu et al., “Converting organosulfur com-pounds to inorganic polysulfides against resistant bacterialinfections,” Nature Communications, vol. 9, no. 1, article3713, 2018.

[32] Y. Liu, H. C. van der Mei, B. Zhao et al., “Eradication ofMultidrug-Resistant Staphylococcal Infections by light-activatable micellar nanocarriers in a murine model,”Advanced Functional Materials, vol. 27, no. 44, article1701974, 2017.

[33] D. Hu, H. Li, B. Wang et al., “Surface-adaptive gold nanopar-ticles with effective adherence and enhanced photothermalablation of methicillin-resistant Staphylococcus aureus bio-film,” ACS Nano, vol. 11, no. 9, pp. 9330–9339, 2017.

[34] H. Zhu, Y. Fang, Q. Miao et al., “Regulating near-infrared pho-todynamic properties of semiconducting polymer nanothera-nostics for optimized cancer therapy,” ACS Nano, vol. 11,no. 9, pp. 8998–9009, 2017.

[35] H. Zhu, J. Li, X. Qi, P. Chen, and K. Pu, “Oxygenic Hybridsemiconducting nanoparticles for enhanced photodynamictherapy,” Nano Letters, vol. 18, no. 1, pp. 586–594, 2018.

[36] D. Xu, C. Zhou, C. Zhan et al., “Enzymatic micromotors as amobile photosensitizer platform for highly efficient on-chiptargeted antibacteria photodynamic therapy,” Advanced Func-tional Materials, vol. 29, no. 17, article 1807727, 2019.

[37] K. Kai, Y. Yoshida, H. Kageyama et al., “Room-temperaturesynthesis of manganese oxide monosheets,” Journal of theAmerican Chemical Society, vol. 130, no. 47, pp. 15938–15943, 2008.

[38] Y. Chen, D. Ye, M. Wu et al., “Break-up of two-dimensionalMnO2 nanosheets promotes ultrasensitive pH-triggered thera-nostics of cancer,” Advanced Materials, vol. 26, no. 41,pp. 7019–7026, 2014.

[39] D. He, X. He, K. Wang et al., “Redox-responsive degradablehoneycomb manganese oxide nanostructures as effectivenanocarriers for intracellular glutathione-triggered drugrelease,” Chemical Communications, vol. 51, no. 4, pp. 776–779, 2015.

[40] Y. Zhang, W. Xiu, Y. Sun et al., “RGD-QD-MoS2 nanosheetsfor targeted fluorescent imaging and photothermal therapyof cancer,” Nanoscale, vol. 9, no. 41, pp. 15835–15845, 2017.

[41] L. Zhao, H. Yang, T. Amano et al., “Efficient delivery of chlorine6 into ovarian cancer cells with octalysine conjugated super-paramagnetic iron oxide nanoparticles for effective photody-namic therapy,” Journal of Materials Chemistry B, vol. 4,no. 47, pp. 7741–7748, 2016.

[42] Q. Chen, L. Feng, J. Liu et al., “Intelligent Albumin-MnO2Nanoparticles as pH-/H2O2-Responsive dissociable nanocar-riers to modulate tumor hypoxia for effective combinationtherapy,” Advanced Materials, vol. 28, no. 33, pp. 7129–7136,2016.

[43] W. Fan, W. Bu, B. Shen et al., “Intelligent MnO2 nanosheetsanchored with upconversion nanoprobes for concurrentpH/H2O2-responsive UCL imaging and oxygen-elevated syn-ergetic therapy,” Advanced Materials, vol. 27, no. 28,pp. 4155–4161, 2015.

[44] S. Lin, H. Cheng, Q. Ouyang, and H. Wei, “Deciphering thequenching mechanism of 2D MnO2 nanosheets towards Aunanocluster fluorescence to design effective glutathione

biosensors,” Analytical Methods, vol. 8, no. 19, pp. 3935–3940, 2016.

[45] G. Yang, L. Xu, Y. Chao et al., “Hollow MnO2 as a tumor-microenvironment-responsive biodegradable nano-platform forcombination therapy favoring antitumor immune responses,”Nature Communications, vol. 8, no. 1, article 902, 2017.

[46] X. Wang, D. Niu, Q. Wu et al., “Iron oxide/manganese oxideco-loaded hybrid nanogels as pH-responsive magnetic reso-nance contrast agents,” Biomaterials, vol. 53, pp. 349–357,2015.

[47] P. S. Stewart and M. J. Franklin, “Physiological heterogeneityin biofilms,” Nature Reviews Microbiology, vol. 6, no. 3,pp. 199–210, 2008.

[48] Y. K. Wu, N. C. Cheng, and C. M. Cheng, “Biofilms in chronicwounds: pathogenesis and diagnosis,” Trends in Biotechnology,vol. 37, no. 5, pp. 505–517, 2019.

[49] K. Mathee, O. Ciofu, C. Sternberg et al., “Mucoid conversion ofPseudomonas aeruginos by hydrogen peroxide: a mechanismfor virulence activation in the cystic fibrosis lung,” Microbiol-ogy, vol. 145, no. 6, pp. 1349–1357, 1999.

[50] C. Nathan and A. Cunningham-Bussel, “Beyond oxidativestress: an immunologist's guide to reactive oxygen species,”Nature Reviews Immunology, vol. 13, no. 5, pp. 349–361, 2013.

[51] Q. Chen, C. Liang, X. Sun et al., “H2O2-responsive liposomalnanoprobe for photoacoustic inflammation imaging andtumor theranostics via in vivo chromogenic assay,” Proceed-ings of the National Academy of Sciences of the United Statesof America, vol. 114, no. 21, pp. 5343–5348, 2017.

[52] J. X. Fan, M. Y. Peng, H. Wang et al., “Engineered bacterialbioreactor for tumor therapy via Fenton-like reaction withlocalized H2O2 Generation,” Advanced Materials, vol. 31,no. 16, article 1808278, 2019.

[53] Y. Jiang, J. Li, Z. Zeng, C. Xie, Y. Lyu, and K. Pu, “Organic pho-todynamic nanoinhibitor for synergistic cancer therapy,”Angewandte Chemie International Edition, vol. 58, no. 24,pp. 8161–8165, 2019.

[54] Q. Miao and K. Pu, “Organic semiconducting agents for deep-tissue molecular imaging: second near-infrared fluorescence,self-luminescence, and photoacoustics,” Advanced Materials,vol. 30, no. 49, article 1801778, 2018.

[55] S. L. Percival, K. E. Hill, D. W. Williams, S. J. Hooper, D. W.Thomas, and J. W. Costerton, “A review of the scientific evi-dence for biofilms in wounds,” Wound Repair and Regenera-tion, vol. 20, no. 5, pp. 647–657, 2012.

[56] G. Nakagami, G. Schultz, D. J. Gibson et al., “Biofilm detectionby wound blotting can predict slough development in pressureulcers: a prospective observational study,” Wound Repair andRegeneration, vol. 25, no. 1, pp. 131–138, 2017.

[57] M. Sønderholm, T. Bjarnsholt, M. Alhede et al., “The conse-quences of being in an infectious biofilm: microenvironmentalconditions governing antibiotic tolerance,” InternationalJournal of Molecular Sciences, vol. 18, no. 12, pp. 2688–2701,2017.

[58] Z. Wang, Y. Zhang, E. Ju et al., “Biomimetic nanoflowers byself-assembly of nanozymes to induce intracellular oxidativedamage against hypoxic tumors,” Nature Communications,vol. 9, no. 1, article 3334, 2018.

[59] J. Hao, G. Song, T. Liu et al., “In Vivo long-term biodistribu-tion, excretion, and toxicology of PEGylated transition-metaldichalcogenides MS2 (M = Mo, W, Ti) nanosheets,” AdvancedScience, vol. 4, no. 1, article 1600160, 2017.

15Research